For many of us, machine learning can be a black box. We call an API, but what happens under the covers is a mystery of calculus and computer science. But understanding some of the concepts that machine learning is founded upon can add yet another tool to our developer toolbox. In this virtual meetup, Matt Dupree will walk through how gradient descent can be used to form predictions in a machine learning model and how to build this in JavaScript.

An Intro to Gradient Descent in Javascript

Gradient descent is the algorithm at the heart of many machine learning problems. In this talk, I’ll introduce the algorithm and code it up from scratch to apply it to a toy linear regression problem on the relationship between videogame metacritic scores and sales.